Academic journal article Journal of International Business Research

Trade Liberalization and Productivity Spillover from Different Foreign Direct Investment Sourcing Origins

Academic journal article Journal of International Business Research

Trade Liberalization and Productivity Spillover from Different Foreign Direct Investment Sourcing Origins

Article excerpt


The study investigates productivity spillovers of different Foreign Direct Investment sourcing origins into domestic firms in 23 Vietnamese manufacturing sectors through horizontal and vertical linkages in two periods: pre and post WTO accession (2004-6, 2007-9). The analysis uses an augmented production function and System GMM estimators to show that the foreign presence by sourcing origin appearing in the same or in downstream industries causes different productivity spillovers into domestic enterprises, and the trade liberalization strongly enforces the spillover process in two directions: (1) Horizontal spillovers are positively improved; (2) Vertical spillovers are strongly diversified. We prove that kind of technology transferred which is more suitable to the local economy' characteristics could drive the spillover absorption through vertical linkages, and domestic firms could actively absorb FDI spillovers through horizontal linkages.

(ProQuest: ... denotes formulae omitted.)


Foreign Direct Investment (FDI) is always attracted by developing countries in hope for more capital for their economic development. While many researchers find evidences on positive FDI impact on one developing country's economic growth and its contribution on taxation, creating more jobs for young population, or supplying wider range of goods and services to the economy, others still argue for its spillover effects on domestic economy (Aitken, Harrison & Lipsey, 1996 and Lipsey & Sjöholm, 2005). On the one hand, it is believed to stimulate the technological progress in the host country. FDI may be used as a vehicle for increasing productivity growth (Bitzer & Görg, 2009). FDI can bring newer technology transfer to developing countries than licensing (Mansfield & Romeo, 1980). In addition, it possibly improves the knowledge and skills of managers or workers, and enhances efficiency and productivity in production and performance. On the other hand, possessing better production technology, managerial skills, export contacts, reputation and good will, FDI is able to force local enterprises to strive in a strong competitive environment and can draw the demand from domestic firms.

However, it is truly difficult to find an existing study on productivity spillovers from different sourcing origins except Javorcik & Saggi (2004). In view of the approach, this study uses firm-level data from the Vietnamese General Statistics Office (GSO) for 23 manufacturing sectors in Vietnamese economy covering the period 2004-2009 to investigate horizontal and vertical productivity spillovers of foreign enterprises sourcing from main traditional counterparts (China, Japan, South Korea, Taiwan, the United States) and associations (ASEAN, Europe). Vietnam has changed to a market oriented economy since 1986. It joined the Association of Southeast Asian Nations (ASEAN) in July 1995 and was fully engaged to the trade liberalization program under ASEAN Free Trade Area (AFTA) in January 1, 2006. After 16 years since applying to participate in the World Trade Organization (WTO) in 1991, Vietnam was accepted to be a full official WTO member in 2007.

Table 1 introduces inward FDI in general and FDI in manufacturing in particular, and FDI by main sourcing origins. In the Vietnamese economy, FDI plays an important role when contributing yearly 16-18% in GDP and foreign firm size is increasing in term of capital when Vietnam is involved more in the regional and the world economies. FDI inflows increased with an average rate from 46% in the period 2004-2005 to 76% in the period 2006-2007, but enormously bumped to 236% in 2008. The largest investors come from Asian countries such as Taiwan, South Korea, Japan, China, Malaysia and Singapore. The United States is a special case when it registered more capital to be the largest investor who occupied 43%» of the whole inward FDI in 2009. While many European countries invest in Vietnam at an average capital level, not much FDI from Africa appear in this developing economy.

The study aims at observing the trade liberalization shock on the productivity spillovers from different sourcing origins by separating two periods pre- and post WTO accession: 20042006 and 2007-2009. Applying two-step System GMM for the production function model helps to give out robust results. By considering the trade liberalization shock in an economy, the analysis firstly contributes evidence that integrating more in the regional and the world trade strongly enforces the spillover absorption process. The sign and magnitude of horizontal spillovers are positively improved, ignoring the serious influence of the world financial crisis in the period 2008-2009. Possibly, severer competition against foreign competitor from trade liberalization cannot surpass the gain of local enterprises from learning, imitating, competing their business activities, and knowledge transfer from labor mobility. Besides, spillovers could be not only driven by a 'negative' absorption so called absorptive capacity but also by a 'positive absorption' with which domestic firms actively select suitable structural characteristics of FDI sources to absorb through demonstration and labor mobility effects.

Moreover, the study makes a further contribution to literature in analyzing the productivity spillovers through vertical linkages from different FDI sourcing origins. It is expected that FDI origins whose the industry technology level are more suitable to the local economy can spill better over to the productivity of domestic firms. Different from investors from ASEAN, Europe, Japan, and South Korea, those from the United States, China, and Taiwan who focused more on selecting low technology industries to begin with after trade liberalization bring positive spillovers. Hence, kind of technology transferred via industry reallocation of investors could affect the potential of spillover absorptive capacity of local firms. This result supports the assumption suggested by Görg & Greenaway (2004).

However, the finding is object to assumptions of 'higher development distance, more imitated technology' of Findlay (1978) or 'higher technology gap between countries, lower spillover potential' of Glass & Saggi (1998). For example, in the period 2004-2006, different from Chinese investors, firms from the United States spilled negatively over domestic enterprises. After the trade liberalization, the appearance of an American enterprise in the same industries or in downstream industries brought the highest positive productivity spillovers. This is contrary to the presence of an ASEAN or a Chinese firm.

The following includes five other sections. Section 2 gives theoretical framework and empirical studies. Then section 3 introduces data, research methodology, and some summary statistics. The results are presented in section 4. Section 5 finally gives conclusion and some discussion.


In fact, a wide range of empirical works have investigated the externalities of inward FDI. Görg & Greenaway (2004) reviewed findings of 45 cases on horizontal and/or vertical productivity spillover of FDI into host developed, transition, and developing economies in the period 1966-2000 and some others on wage, export spillovers. In general, developing economies could suffer negative productivity spillovers but mostly absorb positive spillovers from the appearance of foreign firms in the same industries. However, there were still very few evidences of vertical spillovers. Since the approach of Javorcik (2004) which applied Input-Output (I/O) tables in calculating vertical foreign presence through backward and forward linkages, a large number of papers have deeply analyzed spillover effect of FDI presence in upstream and downstream industries. For example, Javorcik (2004), Kim & Kim (2010) found positive backward productivity spillovers for the cases of Lithuania, Korea, respectively but Bwalya (2006) pointed out negative productivity spillovers for the case of Zambia.

These different results could be explained by a well developed theoretical literature. Once a multinational company (MNC) has established a subsidiary, they are likely to bring along more sophisticated technology, marketing and managerial practices which are possibly spilled over to domestic firms through channels: imitation, skills acquisition, competition and exports (Wang & Blomström, 1992; Aitken & Harrison, 1999). In a more general concept, productivity spillovers can occur through the channels: demonstration, competition, labor mobility, and market stealing effects (Wang & Blomström, 1992, Kokko, 1996, Glass & Saggi, 2002). But in nature, spillovers from FDI are more likely to be vertical than horizontal because MNCs can use ways of protection such as intellectual property, trade secrecy, paying higher wages to prevent labor turnover or locating in countries or industries where domestic firms have limited imitative capacities to begin with (Görg & Greenaway, 2004; Javorcik, 2004).

Referring to the relation between source and host countries, economic theories examine factors affecting the speed of adoption of new technology or driving the degree of horizontal and vertical spillovers. The greater the distance between two economies in terms of development, the more rapidly new technology is imitated (Findlay, 1978). Whereas, Glass & Saggi (1998) concluded that the larger the technology gap between the host and home countries, the lower the quality of technology transferred and the lower the potential for spillovers. In conclusion, Görg & Greenaway (2004) pointed of the absorptive capacity where the spillovers have the potential to raise productivity and exploitation which might be related to the structural characteristics of the host economy. Accordingly, some empirical studies have analyzed the role of both local firm characteristics in the host economy and foreign capital features like firm size, ownership, location, kind of industry, or business orientation (see Attiken & Harrison, 1999; Javorcik, 2004; Le &Pomfret, 2010a).

For empirical works, Javorcik & Saggi (2004) used firm level data for the case of Romania to find that there existed a difference in the magnitude of vertical spillovers associated with multinationals from three regions Europe, America, and Asia. Their findings strongly support the hypothesis from theoretical models of Rodrigues-Clare (1996) and Markusen & Venables (1999) that the share of intermediate inputs sourced by MNCs from a host country is likely to increase with the distance between the host and the source economy. Therefore, they confirmed the role of regional preferential trade agreements which can possibly cause different spillovers of MNCs sourcing from a country in or out of the agreement association. And so, they find evidence of positive productivity spillovers from American and Asian firms but negative spillovers from European firms through vertical linkages.


Data Source

The data used in this study is from the annual enterprise censuses conducted by the GSO. They started from 2000 to survey on 100% of state-owned enterprises and all non-states with employee number no less than 10 in service sectors and 29 manufacturing sectors which are divided into 3 industrial groups: 4 industries in Mining and Quarrying; 2 industries in Electricity, Gas and Water Supply; and 23 industries in Processed Manufacturing (VISC, 1993). The years covered include 2004 through 2009. The number of enterprises increases from a low of 91,755 enterprises in 2004 to a high of 233,236 enterprises in 2009.

This study applies for 23 processed manufacturing industries according to the Vietnamese Standard Industrial Classification (VSIC 1993). Based on the OECD SITC Revision 2_ 1993 (Hatzichronoglou, 1997), these industries are divided into 15 low technology sectors and 8 high technology sectors (see Table 2). According to the GSO, the products of 23 processed manufacturing sectors occupied two third in total manufacturing sectors' and contributed 20.5% in GDP annually in the period 2004-2009. However, FDI inflow to processed manufacturing seriously reduced from 70.5% in 2005 to 17% in 2009. Actually, the registered capital in manufacturing goes from 8.4 trillion USD in 2006 up to 35.7 trillion USD in 2008, but then fell down nearly 8 times in 2009, against 1 .5 times in service.

To analyze the impact of trade liberalization shock when Vietnam joined the WTO, the data set is divided into two periods. After controlling zero and missing values, we have two balanced data sets including 33,150 observations in the period 2004-2006 (28,029 domestic observations and 5,121 foreign observations) and 48,618 observations in the period 2007-2009 (40,507 domestic observations and 8,111 foreign observations). The data sets contain rich information on domestic and foreign ownership, output, sales, assets, employment, location, products, etc. but no direct information of material inputs, except years 2004-2006.

The approach of Javorcik (2004) is applied to calculate horizontal and vertical spillovers from different FDI sourcing origins. Because the surveys are lack of the information of export, we use only backward spillovers representing for vertical effects.

This study also uses the I/O table provided by the GSO (2007) which is the newest one with the dimension of 138 products in order to calculate the backward linkages. The I/O table gives input coefficients in aspect of production technology applied to create products, gross capital formation, final consumptions and exports, and some other indicators. There have been only two versions of the I/O (2000, 2007) so far for the Vietnamese economy. Hence, the version 2007 is used for calculating vertical presence and demand by industry in the whole period 20042009. We assume that the input coefficients are constant over time.

The model and econometric approaches

To meet the study's purpose, a model from augmented Cobb Douglas production function is applied.

In Y^sub ijrt^ = α + ß^sub 1^ ln K^sub ijrt^ + ß^sub 2^ ln L^sub ijrt^ + ß^sub 3^ ln M^sub ijrt^ + ß^sub 4^ Horizontal^sub mjt^ + ß^sub 5^ Backward^sub mjt ^ + ß^sub 6^X + α^sub i^ + ε^sub ijt^

Y^sub ijrt^ is the output which is represented by the sales from the main industry of firm i operating in sector y in region r at time t. The previous studies using the same data source (Le and Pomfret, 2008; Lan, 2008) used output but firms are asked to directly give output at the base year 1994 so the given data could be not correct. Besides, there are much more missing values of output compared to sales. Different from some other studies (Chuc et al., 2008) using total sales, this measure is better to treat the case when the total sales of a firm can come from doing business on other industries, or investing in financial market.

K^sub ijrt^ stands for the capital, defined as the value of fixed assets at the beginning of the year. M^sub ijrt^ , material inputs, are calculated by total expenditure which are equal to total sales minus total profit, adjusted by total wage. There are no specific information of expenditure and wage for the main industry. We assumed total expenditure for products are mostly from payment for materials and labor. Bitzer & Görg (2009) measured materials as the difference between gross output and value added.

Sales, capital, and materials are all deflated by the Producer Price Index for 23 appropriate two-digit manufacturing sectors to get the resulting values at the base year 2004. Labor Lyrt is defined by the number of employees working in the main industry of a firm. Due to lack of data, we cannot apply labor as efficiency units so we accept the same efficiency for a labor working in every enterprise. Javorcik (2004) divided the wage bill by the minimum wage.

Horizontalst captures the presence of foreign firms from country or association m (ASEAN, Taiwan, South Korea, Japan, China, the United States, Europe, and Multi_nations) in sector j at time t, defined by the foreign equity participation (foreign share) averaged over all firms in the sector, weighted by each firm's share in sectoral sales. Foreign share is equal to 100% for a foreign firm without the information of its foreign share. Horizontal from one country or association is measured when 100% of the foreign equity comes from this country or association. We set Hmulti for the presence of foreign multiple shareholders in the industry.

For the purpose of the study, Horizontalmjt could represent for the presence of foreign investors working in low or high technology sectors. That means we can separate horizontals to two types of sectors: low or high technology according to SITC Revision 2 (OECD). Horizontal can cause negative or positive spillover, depending on how local firms overcome the marketing stealing effect from foreign enterprises.


Backward mjt is proxy for the foreign presence in manufacturing from country or association m (or in low or high technology industrial sectors) in downstream industries which are being supplied by sector j at time t. Backward in this model represents for the spillover through vertical linkage, ajk is the proportion of sector y's output supplied to sector k, calculated from the I/O table 2007. The higher appearance of foreign buyers might result a negative or positive productivity effect on local firms.


X includes Indemand, concentration which are controlling variables. The controlling variables are limited to avoid much more missing values.

demandjt stands for demand for intermediates of industry y at time t. One more per cent of the demand is able to force more or less percentage increase in the output depending on the price change in time t. ajk is the I/O coefficient indicating proportion of goody used to produce one unit of good k.


concentration is the Herfindahl index representing the level of industry concentration which is against the competition, equal to sum of all square of relative firm output compared to the whole output in industry j. The increasing index indicates less competition in the industry. That may lead to less productivity growth when lacking of competition effect, or more productivity growth when firms explore resource concentration and effect of increasing return of scale.


From the production function above, many econometric methods could be applied to give different results. In order to obtain robust and consistent coefficients, we must solve the nature problem of error terms. The results from Fixed Effects (FE) estimator will be consistent but those from OLS estimator are both consistent and efficient when the error term is independently and identically distributed. However, observed or unobserved factors from choosing inputs may lead to the correlation between error terms and lagged output, so OLS estimator is no longer consistent (Nickell, 1981). To solve input endogeneity, we use the system GMM which is considered to be consistent and efficient because both lagged differences and lagged levels are used as instruments. Being different from 2SLS, GMM has one more advantage as saving observations when building instrument set whose missing observations are replaced by zeros (Holtz-Eakin, Newey, & Rosen, 1988).

The crucial assumption for the validity of GMM is that the instruments are exogenous. Sargan and Hansen test for joint validity of the instruments is required after GMM estimations. But the Sargan test is inconsistent when the errors are suspected non-sphericity (Roodman, 2006). Arellano and Bond develop a test namely autocorrelation in the idiosyncratic disturbance term. In order to get robust results, the study also uses the recommendation of Roodman (2006) to "collapse" the instrument set and/or limit instruments in confronting the problem of too many instruments. Two-step GMM is applied for Windmeijer (2005) finite-sample correction to the reported standard errors, without which those standard errors tend to be severely downward biased. OLS controlled for industrial, time and regional fixed effects, within group estimator, system GMM estimators are used in this study.

Summary Statistics

As can be seen from Table 3, a foreign enterprise has capital and employee number nearly 50% higher than a local firm on average and it gains higher sales in the main industry. There is not a remarkable difference in domestic and foreign firm size in term of sales, capital, labor, and materials in 2 periods. In detail, a foreign firm's averaged log sales is 10.3, compared to 8.1 of a local firm. Among FDI sources, Htaiwan representing the presence of Taiwanese DI in intra-industries strikingly appears at the largest level 9.7% on average, followed by FDI from Japan, South Korea, ASEAN, and Europe.

In overall, Backwards of all origins are much lower than Horizontals. That means the presence of all investors in downstream industries with the role as customers of domestic firms are much lower than that in the same industries with the role as competitors. The Japanese and the Taiwanese investors purchased more products from domestic enterprises than other sources although their appearance in downstream industries was still very low at 2.9% and 2.5% respectively. It is possible that domestic products are not competitive enough to attract demand of foreign investors from all origins.

After WTO accession in 2007, most of the horizontal and vertical presence of FDI in all sourcing origins slightly increase compared to the previous period except the European and the Japanese investors for the case of horizontal presence, and the American and Taiwanese investors for the case of vertical appearance. Especially, Hchina is from 0.84% in the previous period up to 1.47% in the latter period and Bchina also goes up from 0.18% to 0.32%. The United States, the largest counterpart who occupied over 40%» of the whole inward FDI in 2009 had low presence in the same or downward industries.

Regarding technology level, inward FDI in 2 periods concentrated much in low technology industries. For example, in the period 2004-6, their presence in a low technology industry was 29.8% while that in a high technology industry was just 7.3%. Noticeably, investors appeared more in the same and in downstream low technology industries after Vietnam participated more on trade liberalization. On the contrary, the foreign presence appeared less in intra- high technology industries of domestic firms.

In the manufacturing sectors, FDI inflows increased by 111% from 29.2 trillion VND in the first period to 61.6 trillion VND in the second and the inward capital from all sourcing origins went up. Figure 1 shows a dramatic increase in Japanese DI with the proportion from 13%) up to 29%). The ASEAN and the Taiwanese invested capital also rose up. Nevertheless, this trend was inversed for FDI from countries in Europe. In the post WTO accession, Japan became the largest investors in manufacturing sectors, followed by Taiwan, ASEAN, and South Korea.

Figure 2 presents two groups of sourcing origins: one group has the upward tendency to invest more in low technology industries while the latter has the downward trend. The Chinese, the Taiwanese, the South Korean, and the American investors were moving to do business more in low technology industrial sectors from 2004 to 2009. But this trend was reversed for the cases of ASEAN, Europe, and Japan. By taking the relative horizontal in low/high technology industry yearly, the proportion is 100% when one origin balances the same appearance in high and low technology industries. In detail, FDI from Taiwan only flows to low technology industries because the rates are always larger than 100%. Meanwhile, FDI from South Korea and Japan enters entirely in high technology industries.


Trade Liberalization and Productivity Spillover by Origin

Table 4 shows the results of productivity spillovers through horizontal and vertical linkages by sourcing origin to local firms. Columns 1 through 3 are applied for the period 20046, whereas columns 4 through 6 are for the period 2007-9. The OLS estimation controlling industrial, time, and regional fixed effects so called OLS level gives the results in columns 1 and 4. Columns 2 and 4 reports within or fixed effect estimations. Columns 3 and 6 are for two-step system GMM estimates. Robust standard errors are corrected for finite sample bias. In this study, The GMM estimators only support output and labor endogeneity. Sargan/ Hansen tests show pvalues which are more than 5% so we can conclude that these instruments are exactly valid. The results from system GMM are the most consistent and efficient.

After Vietnam joined the WTO in 2007, the spillovers through horizontal linkages became positive to most of sourcing countries or organizations. Although causing a negative effect in the previous period, American DI triggered the highest spillover to the local economy in the second period. One standard unit increase in foreign presence in the same industry which is equivalent to 100% increase in sales brings 1,680% increase in domestic firms' productivity. This is nearly 4 times higher than the externality effects of Chinese and South Korean DI, but 5 times better than those from European and Taiwanese DI. Robust positive spillovers from the presence of ASEAN firms disappeared when Vietnam integrated more to the regional trade.

Different from horizontal spillovers, vertical spillovers were much diversified. The externalities from American, Chinese and Taiwanese DI in downstream sectors with the role as customers of domestic firms have a good improvement. For example, if the presence of the Chinese buyers went up 1%, productivity of local firms could raise 45.16%. However, the vertical spillovers from European DI were much worse, and those from the Japanese buyers were not significant.

Foreign Direct Investment Spillovers and the Technology Level

The results showed in Table 5 present horizontal and vertical spillovers by technology level of the industry which foreign firms select to begin with. The level of low or high technology of an industry depends on OECD standard industrial classifications. OLS level, fixed effects, and system GMM estimations are applied to 2 periods 2004-2006 and 2007-2009.

From the results in Column 3 and 6, there are robust spillovers from foreign firms which work in low technology sectors through both horizontal and vertical linkages. Any more presence of firms in the same industry, which is at low technology level, could encourage the productivity improvement of local firms. The same trend occurs to the appearance of foreign enterprises in downstream low technology industries. However, the effect is higher under trade liberalization. In detail, the horizontal effect magnitude went up nearly 7 times and the vertical effect magnitude rose about 5 times after Vietnam became an official member of the WTO.

Nevertheless, there was only a significant evidence of positive horizontal spillover of foreign firms at high technology levels for the second period. If the firm presence was 1% higher in a high technology sector, the domestic firms' productivity would be higher via a 1.42% increase in sales. Whereas, the presence of high technology buyers caused a negative effect in the first period but we find no robust results for the period 2007-2009.

What can explain for the different spillover magnitude by sourcing origins?

Spillovers from FDI are more likely to be vertical than horizontal because MNCs can prevent spillovers to their competitors but they are likely to transfer knowledge to local suppliers or buyers (Javorcik, 2004). In this study, this argument is weakened by the affects of trade liberalization. Different from positive horizontal spillovers, there is strong diversification of FDI externalities by sourcing origin through vertical linkages. Reviewing the earliest literature, Findlay (1978) referred to a better FDI externalities sourced from countries which have higher distance in term of development compared to the host country where local firms can imitate the technology. This cannot explain for the cases of South Korean, European, and Japanese DI as foreign buyers do not bring good productivity spillover compared to Chinese DI in the period 2007-2009. This is also not the case of 'higher technology gap between countries, lower spillover potential' (Glass & Saggi, 1998). In fact, the technology gap between Vietnam and the United States is very high compared to the distance from other countries in ASEAN but the American invested enterprises are found to positively spill over the Vietnamese firms through both horizontal and vertical linkages. However, the assumption of Javorcik & Saggi (2004), which concluded that a regional preferential trade agreement may prevent spillover from countries in the association, could make sense. In this study, we find evidence of low spillovers of enterprises coming from countries in ASEAN. But the view point cannot give an explanation to the different effects of firms from China, Japan, and South Korea which all signed a trade agreement in an association with ASEAN, so called ASEAN+3.

In fact, the study supports the assumption of Görg & Greenaway (2004) by analyzing different spillovers depending on kind of technology transferred. Firstly, we realize that after trade liberalization, foreign presence either in low or high industry causes positive spillover. This results in the evidence of positive effects from all sourcing origins except Japan even though Japanese DI appeared at the highest level in the same industry of the local firms (Table 6). There could be the possibility that the Japanese firms have higher capacity of preventing the knowledge leakage to local firms and/or local enterprises find difficulty to learn production and business practices from firms in a high technology industry. American DI, as a special case, brought much higher spillovers compared to the others although it occupied only 8% of the whole inward capital and appears more in high technology industry. From the fact in the business environment in Vietnam, maybe an American firm is the model that a domestic firm tries to imitate and the spillover could go through labor mobility channel, leading to active technology absorption of local firms.

Secondly, concerning the vertical linkages, only the foreign presence in downstream low technology industries fosters a positive productivity spillover. Apparently, the results can explain why FDI from the United States, China, and Taiwan bring good externalities through backward linkages. Furthermore, the remarkable result of Chinese DI between 2 periods might also derive from the fact that they are present more in downstream low technology industries (Table 7). In addition, although the South Korean investors gradually appeared more in low technology industries but like Japan, its presence was still higher in high technology industries so it could be difficult to transfer knowledge to local firms. On the contrary, only investors from ASEAN or Europe appeared more in high technology industries than in low technology industries, and the European investors were faced to a faster downward tendency. Especially, different from all other sourcing origins, they increased their presence in downstream high technology industries. In this case, we find negative productivity spillovers through backward linkages from ASEAN DI in the period 2004-2009, and from Europe DI in both 2 periods.

From the above, we realize a close link from the results in Tables 4 and 5. These precisely show that the different sign and magnitude of different FDI sources through both horizontal and vertical linkages originate from the low technology absorptive capacity of the local firms. However, local firms can actively or negatively absorb the spillover depending on structural characteristics of the host economy. The investors, when selecting types of industry to begin with or changing their orientation by time, possibly affect the spillover channels.


This study investigates the productivity spillovers of inward manufacturing FDI from different sourcing origins. Using firm level data in 23 processed manufacturing sectors which is solved by system GMM estimators for the whole and separated periods from 2004 to 2009 under trade liberalization shock, we give evidence of different horizontal and vertical spillovers by FDI sourcing origin. Those origins are ASEAN, Europe, China, Japan, South Korea, Taiwan, and the United States which are the main traditional counterparts of the Vietnamese economy. The results show that after Vietnam integrates more in the regional economy by completing the AFTA agreement in 2006, or in the world economy by joining the WTO in 2007, American, Chinese, and Taiwanese DI appeared in the same or in downstream industries cause positive productivity spillovers. Meanwhile, other sourcing origins bring good externalities through horizontal linkages, but they do not, or negatively, spill over the productivity of local firms through vertical linkages.

In addition, we argue that the different sign and magnitude of these spillovers cannot be explained by the difference in development or technology between sourcing and host countries, or fully by the regional trade agreement of the host country (Findlay, 1978; Glass & Saggi, 1998; Javorcik & Saggi, 2004). However, the results of the study support the assumption suggested by Görg & Greenaway (2004). That means the absorptive capacity of the local economy and kind of technology that investors select to enter make sense. After the trade liberalization, the Vietnamese firms can absorb much better horizontal and vertical productivity spillovers from American, Chinese and Taiwanese DI when we witness their strong movement toward low technology industry to do business with. We don't see this trend for FDI from ASEAN, Japan, and Europe. More specifically, only South Korean and Japanese DI concentrate more in high technology industries than in low technology sectors. Hence, there is no evidence of good spillovers from them or the spillovers do not match the absorptive capacity of local firms in a developing economy like Vietnam.

However, even there is a tendency toward low technology industries, what can explain for the very different effects between Chinese and American DI for the case of Vietnam? We suggest the concept of "negative/positive absorption". According to Görg & Greenaway (2004), the structural characteristics of the host economy could be related to spillover absorption with the potential to raise productivity and exploitation. We mean that is the negative absorptive capacity. The positive absorption, on the other hand, let the host country to choose which structural characteristics of FDI sources to absorb through demonstration and labor mobility effects.



Anh et al. (2008). Foreign Direct Investment in Vietnam: Is There Any Evidence of Technological Spillover Effects. Retrieved from Development and Policies Research Center (Depocen).

Aitken, B. J. & A. E. Harrison (1999). Do Domestic Firms Benefit from Direct Foreign Investment? Evidence from Venezuela. American Economic Review , 89 (3), 605- 18.

Aitken, B. J., A. E. Harrison & R E. Lipsey (1996). Wages and Foreign Ownership. A Comparative Study of Mexico, Venezuela, and the United States. Journal of International Economics, 40, 345-71.

Bitzer, J. & H. Görg (2009). Foreign Direct Investment, Competition and Industry Performance. The World Economy, Wiley Blackwell, 32(2), 221-33.

Bwalya, S. M. (2006). Foreign Direc Investment And Technology Spillovers: Evidence from Panel Data Analysis of Manufacturing Firms in Zambia. Journal of Development Economics, 81(2006), 514-26.

Cameron, A. C. & P. K. Trivedi (2009). Microeconometrics Using Stata. USA, Texas: Stata Press.

Chuc, Nguyen D. (2008). FDI Horizontal and Vertical Effects on Local Firm Technical Efficiency. Retrieved from Development and Policies Research Center (Depocen).

Findlay, R (1978). Relative Backwardness, Direct Foreign Investment, and the Transfer of Technology: A Simple Dynamic Model. Quarterly Journal of Economics , 92, 1-16.

Glass, A. & K. Saggi (1998). International Technology Transfer and The Technology Gap. Journal of Development Economics , 55, 369-98.

Glass, A. & K. Saggi (2002). Multinational Firms and Technology Transfer. Scandinavian Journal of Economics, 104,495-514.

Görg, H. & D. Greenaway (2004). Much Ado About Nothing? Do Domestic Firms Really Benefit from Foreign Direct Investment? In D. Greenaway, H. Görg & R. Kneller, Globalization and Productivity. UK: Edward Elgar Publishing Limited, Vol. ?, 108-34.

GSO (2010a). The Situation of Enterprises through the Results of Surveys Conducted in 2007, 2008, 2009. Hanoi: Statistical Publishing House.

GSO (2010b). The Input- Output Table for Vietnam in 2007. Hanoi: Statistical Publishing House.

Hatzichronoglou, T. (1997). Revision of the High - Technology Sector and Product Classification. OECD Publishing.

Holtz-Eakin, D., W. Newey & H. Rosen (1988). Estimating Vector Autoregressions with Panel Data. Econometrica, 56, 1371-95.

Javorcik, B. S. (2004). Does Foreign Direct Investment Increase the Productivity of Domestic Firms? In Search of Spillovers through Backward Linkages. American Economic Review , 94 (3), 605-27.

Javorcik, B. S. & K. Saggi (2004). Does It Matter Where You Come From? Vertical Spillovers from FDI and Investor's Nationality. World Bank Policy Research Working Paper.

Kim, H. H. & J. D. Kim (2010). Productivity Spillover Effect of Foreign Direct Investment into Korea. Journal of East Asean Economic Integration, 14(2).

Kokko, A. (1996). Productivity Spillovers from Competition between Local Firms and Foreign Affiliates. Journal of International Development, 8, 517-30.

Lan, N. P. (2008). Productivity spillovers from Foreign Direct Investment: Evidence from Vietnamese Firm Data. Retrieved from School of Commerce, University of South Australia.

Le, Q. H. & R. Pomfret (2008). Technology Spillovers from Foreign Direct Investment in Vietnam: Horizontal or Vertical Spillovers. VDF Working Paper No. 085. Retrieved from http//www. vdf.rg. vn/workingpapers/vdfwp0815.

Lipsey, R.E. & F. Sjöholm (2005). Host Country Impacts of Inward FDI: Why Such Different Answers?. In Blomström, ?., E. Graham & T. Moran, The Impact of Foreign Direct Investment on Development: New Measurements, New Outcomes, New Policy Approaches. Institute for International Economics, Washington D.C, 23-43.

Mansfield, E. & A. Romeo (1980). Technology Transfer to Overseas Subsidiaries by US Based Firms. Quarterly Journal of Economics , 737-50.

Markusen, J. R. & A. J. Venables (1999). Foreign Direct Investment as a Catalyst for Industrial Development. European Economic Review, Elsevier, 43(2), 335-56.

Navaretti, G. B. & A. J. Venables (2004). Multinational Firms in The World Economy. New Jersey: Princeton University Press.

Nickell, S. (1981). Biases in Dynamic Models With Fixed Effects. Econometrica , 1417-26.

Rodriguez-Clare, A. (1996). Multinationals, Linkages, and Economic Development. American Economic Review, 86, 852-73.

Roodman, D. (2006). How To Do xtabond2: An Introduction to "Difference" and "System" GMM in Stata. Center for Global Development. Working Paper Number 103.

Roodman, D. M. (2009). A Note on the Theme of Too Many Instruments. Oxford Bulletin of Economics and Statistics, 135-58.

Wang, J. & M. Blomström (1992). Foreign Investment and Technology Transfer: A Simple Model. European Economic Review, 36,137-55.

Windmeijer, F. (2005). A Finite Sample Correction for the Variance of Linear Efficient Two-Step GMM Estimators. Journals of Econometrics ,25-51.

[Author Affiliation]

Ngoc Thi Bich Pham, University of Kiel and Hoa Sen University

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.